"in one shot" ? Depends a lot on what you mean by that.
It is a little hard to tell what part of your code is setting up the example and which part is the actual attempt to produce the output from that example input.
Let's assume that LOOKUP was your attempt to make datasets named DATASET1 to DATASET4 from SASHELP.CLASS. So something like:
data dataset1 dataset2 dataset3 dataset4;
set sashelp.class;
if findw('Alfred Alice Barbara Carol Henry',Name,' ','st') then output dataset1 ;
if findw('James Jane Janet Jeffrey John',Name,' ','st') then output dataset2 ;
if findw('Joyce Judy Louise Mary Philip',Name,' ','st') then output dataset3 ;
if findw('Robert Ronald Thomas William',Name,' ','st') then output dataset4 ;
run;
And that TEMP is your attempt to make a list of keys to be excluded. So something like:
data temp;
do i=1 to 10;
p=rand('uniform',1,nobs);
set sashelp.class point=p nobs=nobs;
output;
end;
stop;
keep name;
run;
proc sort;
by name;
run;
Does that dataset actually need to have DSNAME variable? Why? Could the same NAME appear in more than one of the datasets? Let's just assume it is not important yet.
So to then split your four input datasets into eight output datasets you might so something like:
data in1 in2 in3 in4 out1 out2 out3 out4;
merge dataset1(in=in1) dataset2(in=in2) dataset3(in=in3) dataset4(in=in4) temp(in=exclude);
by name;
if in1 then if exclude then output out1; else output in1;
if in2 then if exclude then output out2; else output in2;
if in3 then if exclude then output out3; else output in3;
if in4 then if exclude then output out4; else output in4;
run;
Result will look something like this:
179 data in1 in2 in3 in4 out1 out2 out3 out4;
180 merge dataset1(in=in1) dataset2(in=in2) dataset3(in=in3) dataset4(in=in4) temp(in=exclude);
181 by name;
182 if in1 then if exclude then output out1; else output in1;
183 if in2 then if exclude then output out2; else output in2;
184 if in3 then if exclude then output out3; else output in3;
185 if in4 then if exclude then output out4; else output in4;
186 run;
NOTE: There were 5 observations read from the data set WORK.DATASET1.
NOTE: There were 5 observations read from the data set WORK.DATASET2.
NOTE: There were 5 observations read from the data set WORK.DATASET3.
NOTE: There were 4 observations read from the data set WORK.DATASET4.
NOTE: There were 10 observations read from the data set WORK.TEMP.
NOTE: The data set WORK.IN1 has 3 observations and 5 variables.
NOTE: The data set WORK.IN2 has 2 observations and 5 variables.
NOTE: The data set WORK.IN3 has 4 observations and 5 variables.
NOTE: The data set WORK.IN4 has 3 observations and 5 variables.
NOTE: The data set WORK.OUT1 has 3 observations and 5 variables.
NOTE: The data set WORK.OUT2 has 5 observations and 5 variables.
NOTE: The data set WORK.OUT3 has 1 observations and 5 variables.
NOTE: The data set WORK.OUT4 has 1 observations and 5 variables.
NOTE: DATA statement used (Total process time):
real time 0.04 seconds
cpu time 0.04 seconds
So if you know there are FOUR datasets to read in and EIGHT datasets to write out then a few well placed %DO loops will allow you to generate that code pattern. So perhaps something like this:
%macro exclude
(dslist
,excludeds
);
%local i n dsname;
%let n=%sysfunc(countw(&dslist,%str( )));
data
%do i=1 %to &n ;
%let dsname=%scan(&dslist,&i,%str( ));
&dsname._out &dsname._in
%end;
;
merge
%do i=1 %to &n ;
%let dsname=%scan(&dslist,&i,%str( ));
&dsname(in=in&i)
%end;
&excludeds(in=exclude)
;
by NAME;
%do i=1 %to &n ;
%let dsname=%scan(&dslist,&i,%str( ));
if in&i then if exclude then output &dsname._out; else output &dsname._in;
%end;
run;
%mend exclude;
Let's try it:
249 options mprint;
250 %exclude(dataset1 dataset2 dataset3 dataset4,temp);
MPRINT(EXCLUDE): data dataset1_out dataset1_in dataset2_out dataset2_in dataset3_out dataset3_in dataset4_out dataset4_in ;
MPRINT(EXCLUDE): merge dataset1(in=in1) dataset2(in=in2) dataset3(in=in3) dataset4(in=in4) temp(in=exclude) ;
MPRINT(EXCLUDE): by NAME;
MPRINT(EXCLUDE): if in1 then if exclude then output dataset1_out;
MPRINT(EXCLUDE): else output dataset1_in;
MPRINT(EXCLUDE): if in2 then if exclude then output dataset2_out;
MPRINT(EXCLUDE): else output dataset2_in;
MPRINT(EXCLUDE): if in3 then if exclude then output dataset3_out;
MPRINT(EXCLUDE): else output dataset3_in;
MPRINT(EXCLUDE): if in4 then if exclude then output dataset4_out;
MPRINT(EXCLUDE): else output dataset4_in;
MPRINT(EXCLUDE): run;
NOTE: There were 5 observations read from the data set WORK.DATASET1.
NOTE: There were 5 observations read from the data set WORK.DATASET2.
NOTE: There were 5 observations read from the data set WORK.DATASET3.
NOTE: There were 4 observations read from the data set WORK.DATASET4.
NOTE: There were 10 observations read from the data set WORK.TEMP.
NOTE: The data set WORK.DATASET1_OUT has 3 observations and 5 variables.
NOTE: The data set WORK.DATASET1_IN has 3 observations and 5 variables.
NOTE: The data set WORK.DATASET2_OUT has 5 observations and 5 variables.
NOTE: The data set WORK.DATASET2_IN has 2 observations and 5 variables.
NOTE: The data set WORK.DATASET3_OUT has 1 observations and 5 variables.
NOTE: The data set WORK.DATASET3_IN has 4 observations and 5 variables.
NOTE: The data set WORK.DATASET4_OUT has 1 observations and 5 variables.
NOTE: The data set WORK.DATASET4_IN has 3 observations and 5 variables.
NOTE: DATA statement used (Total process time):
real time 0.04 seconds
cpu time 0.04 seconds
Click image to register for webinar
